© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 6798
    Prev
    Next
    INDEX
    Explanations

    verbs or phrases related to restoring, fixing, or healing

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "restore" and its variations, indicating themes of recovery or renewal

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Cube
    -0.71
     Turing
    -0.66
     Hague
    -0.66
     Scouts
    -0.64
     Curve
    -0.63
     crammed
    -0.62
     crowded
    -0.61
     Ide
    -0.61
    balls
    -0.60
     ($)
    -0.60
    POSITIVE LOGITS
     restore
    3.51
     restoring
    2.81
     restores
    2.63
     Restore
    2.49
     restoration
    2.48
     restored
    2.47
     Restoration
    1.81
     regain
    1.65
     rebuild
    1.56
     revive
    1.54
    Activations Density 0.023%

    No Known Activations